🔍 Read the full analysis: The Old Map Is Gone: AI Mathematics, Quantum Computers And The Cryptography Holding Up Finance And Defence on ThorstenMeyerAI.com
Get the latest gadgets delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
A source report says OpenAI published 722 AI-produced mathematical manuscripts on Oct. 6, including results that challenge assumptions about computational speed. Crypto researchers Justin Drake and Vitalik Buterin have raised concerns about AI-discovered algorithms, but no cryptographic scheme is reported broken. The scale and practical relevance of the risk remain unverified.
A reported release of 722 AI-produced mathematical manuscripts has prompted renewed warnings that artificial intelligence could eventually find algorithms capable of weakening cryptography. OpenAI’s results and comments by Ethereum Foundation researcher Justin Drake and Ethereum co-founder Vitalik Buterin have sharpened the discussion, but the source reports no cryptographic scheme has been broken.
According to the source report, OpenAI published the manuscripts on Oct. 6. They were grouped into 372 families and generated by an unreleased internal model from about 4,000 problems, using an average of roughly three hours of ChatGPT Pro compute per result. The reported claims range from results about the Unique Games Conjecture and Hilbert’s tenth problem over the rationals to a zero-free region for the Riemann zeta function. The source does not identify the year of the release.
For cryptographers, the report highlights claims about faster algorithms for computational problems, rather than the famous conjectures. It says computer scientist Scott Aaronson catalogued claims involving integer multiplication and the Fourier transform below n log n, as well as a 3SUM algorithm running in about n^1.9992 time. The source says the 3SUM work appeared in a paper by Virginia Vassilevska Williams and Josh Alman, with a key idea originating from an Anthropic model. These claims are not equivalent to a demonstrated attack on deployed cryptography.
The source also reports that Aaronson noticed cryptography’s absence from the 722 manuscripts and said his sources indicated AI companies were discreetly testing whether internal models could break important protocols. That account is not accompanied here by public test results. Separately, the source says OpenAI withdrew a claimed proof concerning the Hodge conjecture for products of K3 surfaces after a sign error was identified. The episode underlines that AI-generated proofs need independent checking.
The old map is gone: AI mathematics, quantum computers and the cryptography holding up finance and defence
For a decade the plan was simple: elliptic curves doomed by quantum; lattices safe; hashes safe. Nothing has been broken. But a second threat has arrived that doesn’t respect those borders — AI producing new mathematics faster than any human community, against assumptions that are believed, not proven.
Now: on borrowed time — possibly shorter than the quantum countdown suggests.
Now: unproven against AI — and the destination most of the world is migrating to.
Now: reminded estimates move — BSI advised against new deployments on 1 Oct 2026.
Now: safest ground available — not a guarantee.
~n log0.9999999999999 n — a barrier many thought fundamental (OpenAI, claimed)
Overturns a half-century conjecture. Williams & Alman; key idea from an Anthropic model
“Conspicuous by its absence” (Aaronson) — labs reportedly testing crypto “gingerly and discreetly”
ECDSA could break before Q-day, “in the worst case in months not years.” Move funds to never-signed addresses. ~6M BTC sit behind exposed keys.
The new risk is the destination of the migration. Hash-only where possible; “much more paranoid” lattice params; ×10 key sizes long-term. Doesn’t recommend anyone scramble.
“No evidence whatsoever” that elliptic-curve assumptions are close to failing.
Classical breaks could reach “quantum-safe” schemes — but don’t treat a two-year scenario as a date.
Known to IBM and the NSA designing DES (~1974); public via Biham & Shamir (~1990); confirmed by Coppersmith (1994).
Invented at GCHQ — RSA- and Diffie–Hellman-equivalents — and kept secret for over two decades.
No crypto in 722 manuscripts. Found and withheld? Not posed? Posed and failed? Indistinguishable from outside.
Traffic recorded today is decrypted when a break arrives. For secrets that must last 25+ years, a break in 2035 is a break today. A state that finds one won’t announce it — it will mine its archives.
Signatures can be built from hashes. Encryption and key exchange need a trapdoor with structure — lattices, codes or group theory. Defence can only choose which structure, how much margin, how many combined.
Every date was set against quantum hardware forecasts with visible warning. The AI threat offers none.
“ML-KEM everywhere” means starting over if lattices weaken. “We can swap algorithms” doesn’t.
Blockchains show a classical break first — exposed keys and balances are public. Monitor dormant exposed addresses.
Every algorithm, key, certificate, protocol.
PQ + classical, as BSI requires.
Firmware, updates, long-term keys.
Highest sets; evaluate FrodoKEM.
More than one mathematical family; HQC coming.
Swap algorithms without rebuilding.
Forward secrecy, rotation, hidden keys.
Buterin: lost more in botched migrations than in all hacks.
Nothing has been broken, and the sceptics are right that there’s no evidence elliptic curves or lattices are about to fall. But the map has changed: elliptic curves on borrowed time, lattices unproven against AI, codes reminded that estimates move, hashes the safest ground available. For finance, intelligence and defence the answer is the same whichever threat arrives first.The quantum threat comes with a countdown. The AI threat may arrive as a silence — an empty folder where a paper should have been. The winners will be those who can change their algorithms fastest.
Why Crypto Faces a Different Risk
The concern is that AI might find a mathematical shortcut, not that an AI system has already cracked encryption. Many cryptographic systems depend on problems believed to be difficult to solve efficiently, but their security is not generally established by proof that no faster algorithm exists. A new method could change that assessment, although a theoretical improvement would not automatically translate into a practical attack.
This differs from the familiar quantum-computing threat. A sufficiently capable quantum computer running Shor’s algorithm could break RSA and elliptic-curve public-key systems. Progress toward that capability can be tracked through hardware and engineering milestones. A classical algorithm discovered by an AI model, by contrast, could potentially be tested or kept secret without a visible hardware countdown, according to the source’s analysis.
The stakes extend beyond cryptocurrency. Public-key cryptography protects digital communications, financial transactions and government systems. If a widely used mathematical assumption weakened, organisations could face expensive upgrades and uncertainty over data or signatures already protected by affected schemes. These are potential consequences, not evidence that such a failure has happened.
Products Worth Considering
Anker SOLIX C1000 Gen 2 Portable Power Station
Apple Watch SE 3 GPS 44mm Smartwatch
Sonos Era 100 SL Compact, Microphone-Free Speaker – Black
As an affiliate, we earn on qualifying purchases.
Post-Quantum Plans Meet New Questions
Governments and companies have been preparing for quantum computers by moving away from public-key systems considered vulnerable to Shor’s algorithm. The source says the US National Institute of Standards and Technology standardised three post-quantum algorithms in August 2024: ML-KEM for establishing encryption keys, ML-DSA for digital signatures and SLH-DSA, a signature scheme based on hash functions. ML-KEM and ML-DSA are lattice-based.
That migration plan addresses the known quantum threat, but the source argues that AI-driven mathematical research could prompt scrutiny of the assumptions behind newer schemes as well. It does not provide evidence that the standards are compromised. The distinction matters: a possibility worth investigating is not a finding that a standard is unsafe.
Blockchains provide a visible test case because public keys and associated balances can be exposed on-chain. On Oct. 7, Drake urged the industry to plan calmly for “bunker mode,” advising users to move funds to addresses whose public keys had not been exposed. The source estimates that about six million bitcoin are held in addresses with exposed public keys; it gives no further method or date for that estimate.
No Public Break Has Been Shown
The source does not establish that an AI model has found a practical way to break ECDSA, RSA, ML-DSA or another deployed cryptographic scheme. It also does not provide the companies’ test methods, results or independent verification for the reported private evaluations. The warnings therefore remain risk assessments, not proof of an attack.
The status of the 722 manuscripts is also unsettled in the account: it reports a correction and withdrawal after a sign error, but does not say how many other results have been independently checked. It is unclear whether any claimed algorithmic improvements can be implemented at useful scale, what resources they require, or whether they would affect cryptographic systems directly. The year of the Oct. 6 and 7 events is not provided.
Verification and Standards Review
The immediate next step is independent mathematical review of the AI-generated claims and publication of enough detail for researchers to reproduce relevant results. Any claimed cryptographic advance would also need analysis of its complexity, practical resource demands and effect on specific protocols before it could justify changing security guidance.
Organisations already planning post-quantum migrations can continue distinguishing established standards from speculative threats, while researchers examine whether the assumptions behind those standards withstand new algorithmic techniques. For cryptocurrency users, the source presents Drake’s and Buterin’s comments as competing signals about caution and urgency; it reports no general instruction from a standards body or security authority to move funds immediately.
Key Questions
Has AI broken a cryptographic system?
No. The source reports warnings and discreet testing claims, but no demonstrated break of a deployed cryptographic scheme.
What did OpenAI publish?
The source says OpenAI published 722 mathematical manuscripts on Oct. 6, organised into 372 families and generated by an unreleased internal model. It does not specify the year.
How is the AI concern different from the quantum threat?
The quantum concern centres on a sufficiently capable machine running known algorithms against systems such as RSA and elliptic-curve cryptography. The AI concern is that a model could help discover a new algorithm that runs on ordinary computers; the source reports no such cryptographic attack has been established.
Did Vitalik Buterin advise users to move their funds?
The source quotes Buterin saying he did not recommend an immediate scramble to move funds. It also quotes Drake urging the industry to plan for protective measures.
Are post-quantum standards known to be vulnerable?
The source does not report that NIST’s post-quantum standards have been broken. It says the possibility of new mathematical advances has prompted questions about the assumptions behind lattice-based methods.
Source: ThorstenMeyerAI.com
